Rotaviruses, members of the Reoviridae family, have genomes consisting of eleven segments of double-stranded (ds) RNA. In the infectious rotavirus particle, the genome is enclosed within a non-enveloped icosahedron capsid composed of three concentric protein layers. The innermost protein layer is a smooth, thin, pseudo T=1 assembly formed from 12 decamers of the core lattice protein VP2. Tethered to the underside of VP2 layer are complexes comprised of the viral RNA-dependent RNA polymerase (RdRP) VP1 and the RNA-capping enzyme VP3. Together, VP1, VP2, VP3, and the dsRNA genome form the core of the virion. The core proteins function together to transcribe the segmented dsRNA genome, producing eleven capped plus-sense (+)RNAs. The viral RdRP uses the (+)RNAs as templates for the synthesis of the dsRNA genome. Although the RdRP alone can recognize viral (+)RNAs, the polymerase is only active when VP2 is present. The VP2-dependent activity of VP1 provides a means by which genome replication (dsRNA synthesis) can be linked with genome packaging (core assembly). Newly made (+)RNAs pass directly from the RdRP to VP3, an enzyme which introduces m7G caps at the 5'-end of the transcripts through associated guanylyltransferase and methyltransferase activities. VP3 also supports the formation of conduits through the VP2 core lattice protein that are used as exit points for newly made (+)RNAs. An important goal of this project is to characterize the structure and function of the core proteins (VP1, VP2, and VP3). This includes defining the structural interfaces between the proteins, and establishing how these interactions affect and regulate the activities of the proteins. Progress toward this goal is summarized below. (1) RNA interactions with the RNA polymerase. The atomic structure of VP1 was determined by X-ray crystallography though a collaboration with Dr. Steve Harrison's group at Harvard, with details described in a publication appearing in 2008. Briefly, the results showed that the rotavirus polymerase is a compact globular protein with three distinct domains: (i) an N-terminal protruding domain, (ii) a polymerase domain comprised of fingers, palm, and thumb subdomains, and (iii) a C-terminal bracelet domain. Together, the N- and C-terminal domains of VP1 sandwich most of the polymerase domain, creating a cage structure with the catalytic region located within a largely hollow center. Four tunnels connect the surface of VP1 to the catalytic center. These tunnels allow for (i) entry of nucleotides, (ii) entry of single-stranded template RNA, (iii) exit of the dsRNA product or the (-)RNA template, and (iv) exit of (+)RNA transcripts. RNA viruses initiate RNA synthesis through a variety of mechanisms, which as a whole are poorly understood. In the case of rotavirus VP1, a flexible structural element (termed priming loop) near the catalytic site is predicted to stabilize a priming nucleotide in support of RNA initiation. To test this possibility, we expressed and purified a battery of recombinant VP1 proteins with select mutations in the priming loop. Through their analysis, we determined that the priming loop is indeed essential for RNA synthesis. Moreover, we identified which residues of rotavirus priming loop were of particular importance to polymerase function. Further insights into the critical elements and activities of the priming loop are being gained by analysis of mutant VP1 proteins in which the wild type priming loop has been replaced with genetically distinct priming loops of the polymerases of other rotaviruses or other members of the Reoviridae family. The high fidelity of the replication process of RNA viruses demands that viral RdRPs specifically recognize their own template RNAs. Soaks of VP1 crystals with various RNA oligonucleotides has provided insight into the mechanism by which the rotavirus polymerase carries out such recognition. Most notably, such analysis revealed that amino acid residues in the template entry tunnel form base-specific hydrogen bonds with the UGUG portion of a consensus sequence (5'-UGUGACC-3') found at the 3'-end of all rotavirus (+)RNAs. This interaction forms the basis by which the rotavirus polymerase distinguishes its own template RNAs from other RNAs. A remarkable observation is that the extensive network of hydrogen bonds formed between VP1 and its (+)RNA places the 3'-terminal residue of the template past the site required to support RNA initiation. In the presence of VP2, it is believed that conformational changes occur that bring the 3'-terminal residue back into proper register to support initiation. As a part of a program to understand the function of the polymerase, we have produced an extensive set of recombinant baculoviruses that express species of VP1 that include mutations of one or more residues predicted to be involved in template recognition. Mutation of these residues was found to increase the levels of RNA initiation by the polymerase, presumably by weakening the capability of the polymerase to hold the RNA template in the inactive overshot position. These data support the idea that a primary function of template recognition is to stabilize the template in an overshot position, leaving the polymerase in an inactive form that awaits its interaction with VP2 before gaining catalytic activity. This regulatory mechanism assures that the rotavirus genome is not replicated until adequate VP2 is available for supporting core formation and the packaging of newly made dsRNA. (2) VP2-dependent activation of the RdRP. Most human rotavirus isolates can be classified into one of three groups (A, B, or C). Their segmented genome allows rotaviruses to readily exchange genetic material during co-infections. This reassortment process occurs between viruses belonging to the same group, but not for viruses belonging to different groups, for reasons that are unclear. This restriction might reflect the failure of the viral RdRP to recognize and replicate viral RNAs of a different group. To address this possibility, we contrasted the sequences, structures, and functions of RdRPs belonging to rotavirus groups A, B, and C. We found that conserved amino acid residues are located within the hollow center of VP1 near the active site, whereas variable, group specific residues are mostly surface exposed. By creating a three dimensional homology model of the group C RdRP based on atomic structure determined for the group A RdRP, we obtained evidence that these rotavirus RdRPs have nearly identical tertiary folds and share similar mechanisms of recognizing RNA templates. Consistent with their structural analysis, we determined that recombinant group A and C RdRPs are capable of replicating one anothers RNA templates in vitro. However, the activity of both RdRPs is strictly dependent on the presence of their cognate VP2 core lattice protein. That is, the group A RdRP has activity in the presence of group A VP2, but not group C VP2, and vice versa. Thus, the reassortment restriction between rotavirus groups may reflect the inability of their replication proteins to function together in support of genome replication. (3) Expression of the RNA capping enzyme. Previously, the expression of recombinant VP3 at levels necessary to provide adequate amounts for pursuing X-ray crystallography has not been achieved. To approach this problem, cDNAs were prepared to the VP3 genes of three genetically-distinct prototype strains of rotavirus. The cDNAs were made such that the codons of the VP3 open reading frame were optimized for expression in insect cells. Analysis of recombinant baculoviruses expressing these cDNAs showed that they produced large amounts of soluble VP3. Experiments are underway to optimize purification protocols and to establish functional assays of the recombinant protein